1. Design and synthesis of novel Grb2 SH2 domain binding antagonists. Based on the unique amino acid recognition sequence and three dimensional structure of the Grb2 SH2 domain, potent, synthetic, low molecular weight antagonists of Grb2 SH2 domain binding were designed and synthesized in collaboration with Dr. Terry Burke (NCI/CCR/Chemical Biology Laboratory). Important innovations made during the prior report period included stabilizing the phosphotyrosyl mimetic group against constitutive cytosolic tyrosine phosphatase activity and systematic substitution of the surrounding SH2 domain recognition motif to achieve low nanomolar affinity synthetic compounds. Structural improvements during this report period were focused on macrocyclization to stabilize the required beta-turn conformation. Current and future efforts toward Goal 1 address improved routes of synthesis, modifications to improve biological half life and bioavailability in animals, and the extension of our Grb2 approach to the SH2 domains of other critical oncogenic mediators. 2. Defining the roles of Grb2 in oncogenesis and metastasis. Prior work toward this goal showed that our Grb2 SH2 domain binding antagonists potently blocked receptor TK-Grb2 interaction and growth factor-stimulated motility and matrix invasion by several tumor cell lines, including those derived from RCC, prostate and bladder cancer, as well as angiogenesis in vitro and in vivo. The potent effects of our Grb2 SH2 domain binding antagonists on cell motility and matrix invasion prompted us to focus in this report period on models of tumor metastasis for the preclinical development of these compounds in animals. Using induced and spontaneous experimental metastasis mouse models in conjunction with bioluminescence to quantitatively characterize metastasis, a significant reduction in metastatic burden was demonstrated using our prototypical Grb2 SH2 domain binding antagonist administered ex vivo to the implanted cells. This route of administration used minimal amounts of drug and because of the short half-life of our prototype inhibitor, the results also revealed that the events required for successful colony formation are likely to occur within the first 24 h after tumor cell implantation. Our work also showed that HGF-induced Met-Grb2 binding and Grb2-FAK interaction were significantly disrupted by these agents, resulting in dramatically reduced focal adhesion and lamellipodia formation in an aggressive human prostate cancer cell line. These results further defined the critical role for Grb2 in regulating growth factor-stimulated cell motility in normal and transformed cells, and identified Grb2 binding partners that could be targeted for inhibition independently of Grb2. Current and future investigations within Goal 2 exploit unique tools developed in Dr. Terry Burkes lab under Goal 1 that are derivatives of the Grb2 SH2 domain binding antagonists: a biotin conjugate that can be captured from biological samples using immobilized streptavidin and a fluorescent dye conjugate for real-time quantitative imaging. These reagents will be used to identify Grb2 binding partners in GU tumor-derived cell lines with defined genetic defects and to further characterize the functions of Grb2 in focal adhesion and lamellipodia formation. Interestingly, Grb2 has been found in human urinary exosomes;we have confirmed this observation and found Grb2 in human plasma and in the conditioned media of cultured cells. Extracellular Grb2 is stable and mass spectrometry analysis of biotin-Grb2 SH2 domain antagonist-captured proteins shows association with known interactors as well as new associations unique to the extracellular milieu, some likely to be mediated by the Grb2 SH3 domains. 3. Preclinical development of SH2 domain antagonists as anti-cancer drugs. Because Grb2 SH2 domain binding antagonists were found to be highly effective in blocking cell motility, models of tumor metastasis were sought for the preclinical development of these compounds in animals. As mentioned above, we used a murine melanoma cell line in an experimental metastasis model and a human prostate adenocarcinoma cell line in a spontaneous metastasis model, in conjunction with bioluminescence technology to quantitate metastasis, and found a significant reduction in metastatic burden could be achieved using a prototypical Grb2 SH2 domain binding antagonist. These results demonstrated that it was possible to specifically target the spread of solid tumors using small molecules and it implied a critical role for the Grb2 SH2 domain in that process. Several important steps remain in the preclinical development of Grb2 SH2 domain binding antagonists as anti-cancer drugs. These compounds are among the first and most successful antagonists of protein-protein interactions, but defining target selectivity and the molecular basis of therapeutic effects are important steps toward their development and use. Current and future efforts will also focus on drug formulation, conventional administration routes and pharmacokinetics, toxicology and additional drug efficacy studies. Initial pharmacokinetic studies of a Grb2 SH2 domain binding antagonist in mice indicate that while it had a relative short biological half-life, target drug doses could be maintained continuously using implanted osmotic minipumps. Further studies of drug formulation in lipophilic nanoparticles to improve half-life and bioavailability are also underway. Thus far, no toxicity has been observed in mice treated for four weeks with doses designed to exceed the ED50 by 100-fold, and further toxicological studies are in progress. Finally, the development of pharmacodynamic markers of drug action in intact animals represents another high priority in the preclinical development of our Grb2 SH2 domain antagonists. We have performed a preliminary analysis of global gene expression changes associated with drug treatment which will be used to identify a molecular signature of drug effect. Results from these studies suggest that N-cadherin, implicated in the increased invasiveness of many cancers, is among this set of potential pharmacodynamic markers.